DOWNLOAD: © A.J. Arnfield, 2006

advertisement
How to do Urban Climatology:
a Personal View of the Legacy
of Professor Tim Oke
A. John Arnfield
Geography, The Ohio State University, USA
Geography, Earth & Environmental Sciences,
University of Birmingham, UK
Outline




Definitions
Preface: some preliminary remarks
The Major Themes: a brief review of the
framework and major findings of a subset of
modern urban climatology from the
perspective of Oke’s work
The Oke Contribution: a personal
assessment the Oke contribution to the
vigour, rigour, visibility, scope and content of
urban climatology
Definitions




Following Oke (2006), I will use the term urban climate to include “the
study of meteorological processes, atmospheric phenomena, and the
longer term amalgam expression of these as climates in areas that
have undergone urban development.” Likewise with derived words
such as “urban climatology” or “urban climatologist”.
In the following, I will include within the Oke opus any work on which his
name appears in the author list and will make no attempt to differentiate
Oke’s work from that of his co-authors.
I am confining consideration to the published, peer-reviewed literature
in urban climate and will not include reports, non-refereed articles,
proceedings etc.
Finally, I shall in places use the following abbreviations for energy
budget components





Q*
QH
QE
ΔQS
QG
flux density of net radiation
turbulent flux density of sensible heat
turbulent flux density of latent heat
storage heat flux density (in local scale budget)
substrate heat flux density (in budget for a simple surface)
Preface [1]



Timothy R. Oke: PhD (McMaster University),
Geography, 1967
Dissertation: Super-surface Minimum Temperatures
at Night
First publications in urban climatology (as Assistant
Professor, McGill University):


1968. Some results of a pilot study of the urban
climate in Montreal. McGill Climatological Bulletin 3,
36-41
1969. Towards a more rational understanding of the
urban heat island. McGill Climatological Bulletin 5, 121
Preface [2]

Themes in Oke’s first publications:







City effects in changing radiation and energy budgets
Vertical extent of the city influence on the atmosphere
3D structure of the urban heat island and its dependence
on meteorological conditions
Meticulous attention to experimental design, observation
and sampling error and to theory-measurement coupling
Emphasis on the study of process with the objective of
establishing a sounder physical basis for urban climates
Urban-rural contrasts in surface and substrate properties
Skepticism concerning some contemporary theories of
heat island causality, including urban-rural thermal
properties, supposed reduced urban evaporation rates &
pollution effects on the incoming longwave irradiance
Preface [3]

Themes in Oke’s first publications:




The problem of determining the heat storage flux in
urban areas
Issues in adapting standard micrometeorological
methods of eddy flux measurement to the city, including
spatial heterogeneity, definition of the urban “surface”,
flux constancy with height, equality of eddy diffusivities
and advection (eddy covariance deemed most desirable)
The necessity to develop models of urban climate and its
formative processes
Importance of atmospheric energy budgets (fluxdivergence)
The Major Themes [1]



I will present a skeleton outline of the structure and workings of
the urban climate system, as we understand it today
I will confine topical coverage to Oke’s primary areas of
concentration – urban physical climatology, urban
micrometeorology, urban mesoclimatology and urban boundary
layer climatology
I will emphasize Oke’s contributions (conceptual frameworks,
findings, theories, methods and models), i.e. ideas:
 initiated by Oke and pursued in his work and by others working in
the field
 adopted from the work of other urban climatologists but
enhanced, expanded and advanced by Oke
 adopted from other areas of the atmospheric science and applied
in the context of the urban environment by Oke
The Major Themes [2]
The Urban Climate System
The distinction between the Urban
Boundary Layer and the Urban Canopy
Layer
• Oke (1976)
• Oke (1982)
• Oke (1988b)
• Roth, Oke & Emery (1989)
• Rotach et al. (2005)
More than a matter of scale: climates owe
their genesis to different processes in the
UBL and UCL, a different theoretical
framework is needed for each and work
on their coupling is imperative
• Oke (1976)
• Oke (1982)
• Nakamura & Oke (1988)
UBL
UCL
Coupling of the UCL and the UBL
• Nunez & Oke (1976)
• Oke (1976)
• Nunez & Oke (1977)
• Oke (1982)
• Nakamura & Oke (1988)
• Salmond, Oke, Grimmond, Roberts &
Offerle (2005)
The Major Themes [3]
The Urban Climate System

UBL Subdivisions


Mixed Layer (or Outer
Urban Boundary Layer)
Surface Layer


Roughness Sublayer
Inertial Sublayer (or
Constant Flux Layer)
Capping Inversion
UBL
UCL
The Major Themes [4]
The Urban Climate System

Surface Layer
Capping Inversion
Mixed Layer
Surface
Layer
UCL
Mixed Layer + Surface Layer = UBL
The Major Themes [5]
The Urban Climate System

Roughness Sublayer







Roth, Oke & Steyn (1989)
Oke, Cleugh, Grimmond,
Schmid & Roth (1989)
Schmid, Cleugh, Grimmond
& Oke (1991)
Roth & Oke (1993)
Grimmond & Oke (2002)
Grimmond, Salmond, Oke,
Offerle & lemonsu (2004)
Rotach et al. (2005)
Capping Inversion
Mixed Layer
Constant
Flux
Layer
Roughness
Layer
UCL
Constant Flux Layer + Roughness Layer = Surface Layer
Mixed Layer + Surface Layer = UBL
The Major Themes [6]
The Urban Climate System

Inertial Sublayer (constant
flux layer)







Roth, Oke & Steyn (1989)
Oke, Cleugh, Grimmond,
Schmid & Roth (1989)
Schmid, Cleugh, Grimmond
& Oke (1991)
Roth & Oke (1993)
Grimmond & Oke (2002)
Grimmond, Salmond, Oke,
Offerle & Lemonsu (2004)
Rotach et al. (2005)
Capping Inversion
Mixed Layer
Constant
Flux
Layer
Roughness
Layer
UCL
Constant Flux Layer + Roughness Layer = Surface Layer
Mixed Layer + Surface Layer = UBL
The Major Themes [7]
Turbulence, Micrometeorological Theory & Measurement

Absence of similarity in transfer
efficiencies for different entities






Oke (1988b)
Roth, Oke & Steyn (1989)
Roth & Oke (1993)
Roth & Oke (1995)
Kanda, Moriwaki, Suzuki, Roth &
Oke (2000b)
Capping Inversion
Mixed Layer
Constant
Flux
Layer
Applicability of Monin-Obukhov
Theory in the Urban UBL





Oke (1988b)
Roth, Oke & Steyn (1989)
Oke, Cleugh, Grimmond, Schmid
& Roth (1989)
Roth & Oke (1993)
Kanda, Moriwaki, Suzuki, Roth &
Oke (2000b)
Roughness
Layer
UCL
Constant Flux Layer + Roughness Layer = Surface Layer
Mixed Layer + Surface Layer = UBL
The Major Themes [8]
Turbulence, Micrometeorological Theory & Measurement

Turbulence spectra, cospectra
& other statistics over
suburban terrain







correspondence to reference
spectra
no evidence for spatial scales
associated with the UCL
dissipation of TKE relatively
small compared to reference
cases
Roth, Oke & Steyn (1989)
Oke, Cleugh, Grimmond,
Schmid & Roth (1989)
Roth & Oke (1993)
Roth & Oke (1995)
The Major Themes [9]
Turbulence, Micrometeorological Theory & Measurement

Coherent structures in
urban boundary layer
(Marseille)


Salmond, Oke, Grimmond,
Roberts & Offerle (2005)
Aerodynamic properties of
urban areas at the local
scale




Grimmond, King, Roth &
Oke (1998)
Grimmond & Oke (1999b)
Kanda, Moriwaki, Suzuki,
Roth & Oke (2000a)
Kanda, Moriwaki, Roth &
Oke (2002)
The Major Themes [10]
Turbulence, Micrometeorological Theory & Measurement

Measurement issues for urban
surfaces

measurement height







Oke (1988b)
Roth, Oke & Steyn (1989)
Oke, Cleugh, Grimmond,
Schmid & Roth (1989)
Schmid, Cleugh, Grimmond &
Oke (1991)
Roth & Oke (1993)
Rotach et al. (2005)
validity of assumptions for eddy
flux measurement techniques





Oke (1988b)
Roth, Oke & Steyn (1989)
Roth & Oke (1993)
Roth & Oke (1995)
Kanda, Moriwaki, Suzuki, Roth
& Oke (2000b)
The Major Themes [11]
Turbulence, Micrometeorological Theory & Measurement

Measurement issues for urban
surfaces

spatial/time sampling








Roth, Oke & Steyn (1989)
Oke, Cleugh, Grimmond,
Schmid & Roth (1989)
Schmid, Cleugh, Grimmond &
Oke (1991)
Roth & Oke (1993)
Roth & Oke (1995)
Kanda, Moriwaki, Suzuki, Roth
& Oke (2000a)
Kanda, Moriwaki, Roth & Oke
(2002)
Mestayer et al. (2005)
The Major Themes [12]
The UCL Urban Heat Island
UCL heat island spatial
structure



Oke & East (1971)
Oke (1982)
Runnalls & Oke (2000)
The Major Themes [13]
The UCL Urban Heat Island

UCL heat island
dynamics






Oke & East (1971)
Oke & Maxwell (1975)
Oke (1981)
Oke (1982)
Runnalls & Oke (2000)
Brunt cooling curve



Oke & Maxwell (1975)
Nunez & Oke (1976)
Oke (1981)
The Major Themes [14]
The UCL Urban Heat Island

Oke’s Proposed UCL UHI Causes (Oke, 1982)







Increased absorption of solar radiation (canyon geometry)
Increased long-wave counter-radiation (air pollution)
Decreased net long-wave radiation loss (canyon geometry)
Anthropogenic heat sources (releases into canyon air)
Increased sensible heat storage in urban fabric (materials –
thermal admittance)
Decreased evapotranspiration (materials - vegetated area
reduction and impervious surfaces)
Decreased vertical mixing (canyon geometry – shelter)
The Major Themes [15]
The UCL Urban Heat Island

ΔTu-r dependence on wind speed






Oke and East (1971)
Oke (1973)
Oke (1976)
Oke (1982)
Runnalls & Oke (2000)
ΔTu-r dependence on cloud cover





Oke & East (1971)
Oke and Maxwell (1975)
Oke (1982)
Oke, Johnson, Steyn & Watson
(1991)
Runnalls and Oke (2000)
Urban heat island
intensity
ΔTu-r = Tu - Tr
The Major Themes [16]
The UCL Urban Heat Island

ΔTu-r dependence on urban
material thermal admittance





Oke & Maxwell (1975)
Oke (1981)
Oke (1982)
Oke, Johnson, Steyn & Watson
(1991)
ΔTu-r dependence on canyon
aspect ratio or SVF






Oke (1976)
Oke (1981)
Oke (1982)
Oke (1988b)
Johnson, Oke, Lyons, Steyn,
Watson & Voogt (1991)
Oke, Johnson, Steyn & Watson
(1991)
The Major Themes [17]
The UCL Urban Heat Island


Modelling the heat island –
SHIM
 Johnson, Oke, Lyons,
Steyn, Watson & Voogt
(1991)
 Oke, Johnson, Steyn &
Watson (1991)
Simultaneous modelling of
the role canyon geometry
on street surface L*,
thermal admittance,
anthropogenic heat and
enhanced urban counter
radiation due to pollution
and cloud cover
The Major Themes [18]
The UCL Urban Heat Island

The role of the rural: rural
thermal admittance, as
controlled by soil moisture



Oke, Johnson, Steyn &
Watson (1991)
Runnalls & Oke (2000)
The role of the rural:
complete energy budget
approach (including “cool
islands”)


Grimmond, Oke & Cleugh
(1993)
Rotach et al. (2005)
The Major Themes [19]
The Surface Urban Heat Island

Characteristics of satelliteobserved surface UHIs



most pronounced during the
day and weakest at night
spatial distribution of
temperature closely tied to land
use: peaks often associated
with areas of large flat roofs
and ground level paved
surfaces (rather than tall
buildings)
correlation with UCL (air
temperature) UHIs often very
weak (frequently negative)
The Major Themes [20]
The Surface Urban Heat Island

Roth, Oke & Emery (1989)
 What is the nature of the urban surface as “seen” by a
remote sensor?
 How do remote sensor detected radiative temperatures
relate to the true air-urban interface?
 What is the relationship between satellite-derived UHIs and
those measured in air?
 How appropriate is thermal remote sensing data as input to
urban climate models?
The Major Themes [21]
The Surface Urban Heat Island








Nakamura & Oke (1988)
Roth, Oke & Emery (1989)
Voogt & Oke (1997)
Voogt & Oke (1998a)
Voogt & Oke (1998b)
Voogt & Oke (2003)
Soux, Voogt & Oke (2004)
Mestayer et al. (2005)
The Major Themes [22]
The Surface Urban Heat Island

Issues in remote sensing of the UHI



Nadir views oversample the
horizontal facets of the urban surface
and undersample vertical ones.
Other sensor views give biased
samples of the true urban surface
temperature, depending on surface
geometry and the relationship of the
sun, city and sensor platform to one
another.
“Effective anisotropy”: can lead to
temperature differences of similar
order to those associated with
atmospheric corrections and much
larger than emissivity effects,
especially in zones with a regular
block structure, large areas of vertical
surface and with sunlit/shaded
contrasts.
Problems in detecting true surface
temperature may enhance of reduce
the UHI signal.
The Major Themes [23]
The Surface Urban Heat Island

Issues in remote sensing of the UHI



It is essentially futile to expect any
simple relationship between
remotely-sensed surface temperature
and air temperatures in cities.
The remotely-sensed surface
temperature will not, in general, equal
the “complete surface temperature”:
neither, in general, will it equal to
“aerodynamic temperature” required
for use in energy budget model
calculations.
Modelling what a sensor “sees” of an
urban area is a non-trivial problem
but can yield information on what
types of sensing yield a “good”
sample: a “good: sample is one in
which the proportions of the surface
elements observed is similar to the
mix contributing to the complete
surface temperature.
The Major Themes [24]
Urban Canopy Energy Budget and Climate


The urban canopy
comprises myriad surfaces
and forms arranged in
complex 3D configurations
Typical repeating structure
– the “urban canyon”






Nunez & Oke (1976)
Nunez & Oke (1977)
Nunez & Oke (1980)
Nakamura & Oke (1988)
Rotach et al. (2005)
Oke first to formally define
the urban canyon?
The Major Themes [25]
Urban Canopy Energy Budget and Climate


Vancouver (north-south)

by day, QH is the dominant sink for
Q* for all facets, each exhibiting a
unique and complex diurnal pattern
driven by both incoming and
reflected solar radiation

by night, the budget simplifies to
ΔQS ≈ Q*

expressed relative to the canyon
top, the budget is similar to that of
a horizontal surface (except for a
shorter period of positive Q*) with:

QH ≈ 0.64Q*, QE ≈ 0.10Q* and
ΔQS ≈ 0.26Q* (day)

ΔQS ≈ Q*, QH ≈ -QE
Basel

QH through the canyon top
(measured by scintillation) similar
in pattern to a horizontal surface,
upward-directed at night & slightly
smaller in magnitude than QH from
the adjacent roof
The Major Themes [26]
Urban Canopy Energy Budget and Climate

Vancouver (north-south)




Down-canyon advection and heat transport by the mean
flow through the canyon top are likely to be significant in
the air volume energy budget
Divergence of L* occurs during the night both across and
along the canyon
Radiative cooling rates in the early nocturnal period exceed
actual rates (suggesting QH convergence) but decrease
later
Radiative cooling within the canyon is less than that at roof
level and less than typical rural rates
The Major Themes [27]
Urban Canopy Energy Budget and Climate

Kyoto (east-west)






Canyon air is generally well-mixed, except within 0.5-1.0 m
of the facets
Air temperature structure strongly conditioned by solar
irradiation patterns
Roof air is slightly warmer than canyon air by day and
cooler by night but differences are small
Evidence of a vortex within the canyon with a “corkscrew”
pattern with ambient diagonal winds
Cross-canyon flow gives the strongest temperature
contrasts within the canyon and along-canyon flow the
least
The air within the canyon is unstable at all times, including
the night
The Major Themes [28]
Urban Canopy Energy Budget and Climate

Irrigated lawn (Oke, 1979)



Q* peaked by day and only
moderately negative at
night (skyline obstruction)
QE dominates the energy
budget and exceeds Q* for
much of the afternoon and
evening (inversion and
downward-directed QH)
QE enhanced by sensible
heat advection at the microscale from surrounding
warm impervious surfaces
The Major Themes [29]
Urban Canopy Energy Budget and Climate

Urban parks (Spronken-Smith & Oke, 1998, 1999)






Park “cool islands” based on surface temperature may be large with
temperatures variable by park type.
Cool islands defined with respect to air temperature are more muted,
rarely more than 3oC cooler than built-up surroundings, into which
lower temperatures may be advected about one park width
Scale modelling suggests that, as with canyons, both SVF and
thermal admittance of materials are contributory to the degree of
coolness of a park
Edge effects (warmer surface) extend about 2.2X-3.5X the height of
the park border into the park
A park large enough to build up a cool air pool may initiate a
centripetal circulation that will extend the cooling into the adjacent
built-up area
Moist parks with tree cover may be cool by day but may exhibit
retarded cooling at night (thermal admittance and SVF): evaporation
may indirectly play a role in cooling parks at night by starting cooling
from a lower temperature at sunset
The Major Themes [30]
Urban Canopy Energy Budget and Climate

Irrigated urban park (SpronkenSmith, Oke & Lowry, 2000)




Greatest evaporative
enhancement at upwind edge
(leading edge effect) with QE >
Q*
In park centre, QE smaller but
also > Q* in afternoon (oasis
effect)
QH directed towards the
surface (inversion)
Thermostat-effect: irrigated
park behaves analogously to a
wet leaf in a hot environment
(Sacramento) – surface
temperature never gets more
than a couple of degrees
above air temperature and is
below it for much of the day
The Major Themes [31]
Local-Scale Urban Surface Energy Budgets

Residential-Suburban Sites











Kalanda, Oke &
Spittlehouse (1980)
Oke (1982)
Oke & McCaughey (1983)
Cleugh & Oke (1986)
Grimmond & Oke (1986)
Oke (1988b)
Schmid, Cleugh, Grimmond
& Oke (1991)
Grimmond, Oke & Cleugh
(1993)
Roth & Oke (1994)
Grimmond & Oke (1995)
Grimmond & Oke (2002)
Q*
QH
QE
QS
DAY
NIGHT
The Major Themes [32]
Local-Scale Urban Surface Energy Budgets

Residential/Suburban Sites



Very little spatial variability in Q* over suburban areas
QH usually the largest daytime sink, often asymmetric around solar noon,
showing a long “tail” to midnight
QE larger than might be expected based on vegetated fraction (although not
as high as early studies suggested), perhaps:








maintained by piped water supply
enhanced by micro-scale advection and oasis effects for small moist surface
patches
augmented under appropriate synoptic conditions by mesoscale advection acting
through entrainment of dry air brought about by roughness-induced mechanical
turbulence (strong coupling with PBL)
“Rebound” in evaporation rate during extended dry periods after 4-6 days,
attributable to perceptual lag of homeowners and correlation between piped
water supply use and evaporation rates
Correlation between Bowen ratio and area irrigated.
Marked variation in partitioning between the turbulent fluxes on a day-to-day
basis – attributed to synoptic variability and PBL coupling
ΔQS exhibits marked hysteresis (flux larger in morning than afternoon)
Afternoon decline in ΔQS supports the QH “tail” until late in the evening
The Major Themes [33]
Local-Scale Urban Surface Energy Budgets

City Centre & Industrial Sites








Oke, Zeuner & Jauregui (1992)
Oke, Spronken-Smith,
Jauregui & Grimmond (1999)
Grimmond & Oke (2002)
Grimmond, Salmond, Oke,
Offerle & Lemonsu (2004)
Mestayer et al. (2005)
Lagouarde, Irvine, Bonnefond,
Grimmond, Oke, Salmond &
Offerle (2005)
Offerle, Grimmond, Fortuniak,
Klysik & Oke (2005)
Rotach et al. (2005)
Q*
QH
QE
QS
DAY
NIGHT
The Major Themes [34]
Local-Scale Urban Surface Energy Budgets

City Centre & Industrial Sites






Daytime budget dominated by sensible heat fluxes (Q H and ΔQS each dominate in
different cities)
QH asymmetric with a peak after solar noon and long “tail” through afternoon and
evening, to upward-directed flux at night
Scintillometry generally confirms QH magnitudes but with smoother diurnal regime
attributed to better spatial sampling compared to eddy covariance method
ΔQS is far larger than typical for suburban sites (and for most rural sites), peaks before
solar noon, declines through the afternoon, goes negative before the net radiation, and
provides a source of heat to maintain an upward directed sensible heat flux at night
ΔQS exhibits marked hysteresis (flux larger in morning than afternoon)
QE:






often larger than expected on the basis of restricted greenspace but usually the smallest energy
budget component
seems to depend on the fraction of vegetation and open water present (which can be large in
some cities)
may be enhanced by anthropogenic sources, such as street washing, piped water leaks etc
may persist at low rate throughout the night
At higher latitudes, anthropogenic heat may be relatively significant during winter when
Q* is small and may maintain turbulent fluxes in excess of the net radiation
Variability in magnitudes of and partitioning between the turbulent fluxes between years
(precipitation), seasonally and on a day-to-day basis (synoptic variability enhanced by
good PBL coupling)
The Major Themes [35]
Local-Scale Urban Heat Storage (ΔQS) Modelling







Oke, Kalanda & Steyn (1981)
Oke and Cleugh (1987)
Grimmond, Cleugh & Oke (1991)
Roth & Oke (1994)
Grimmond & Oke (1999a)
Grimmond & Oke (1999c)
Grimmond & Oke (2002)
The Major Themes [36]
Local-Scale Urban Heat Storage (ΔQS) Modelling




ΔQS is not a negligible term in the energy
budgets of urban areas (17-58% of Q*).
In absolute terms, ΔQS tends to be
largest for downtown sites and smallest
for residential areas.
There exists no feasible way of directly
measuring ΔQS: may be parameterized in
terms of Q* or found as the energy
budget residual.
Parameterizations:

linear regression between ΔQS and Q*

inclusion of hysteresis between ΔQS
and Q* by including both Q* and
∂Q*/∂t as predictor variables

Objective Hysteresis Model: as above
but with coefficients determined by
weighting those for individual surface
types (canyons, roofs, lawn, etc)
according to surface fraction based on
a plan area or 3D area survey
The Major Themes [37]
Local-Scale Urban Heat Storage (ΔQS) Modelling

OHM





Generally performs well
Tends not to reproduce large releases of heat in the
afternoon for some sites
Tends to underestimate the hysteresis behaviour of urban
surfaces
Does not capture all hour-to-hour variation, possibly
because it is unable to account for short-term changes in
moisture conditions, wind speed variations, synoptic
conditions etc, which will affect the turbulent fluxes and,
hence, ΔQS
Would benefit from availability of a greater range of
coefficients for more urban surface types, especially those
that undergo seasonal changes in characteristics
The Major Themes [38]
Local-Scale Urban Energy Budget Modelling

NARP (Net All-Wave Radiation Parameterization): uses commonly
available meteorological data, plus local scale emissivity and albedo
information to estimate Q* for Lodz, Chicago and Los Angeles


Offerle, Grimmond & Oke (2003)
A test of the ability of three models (Myrup, 1969; Ackermann, 1977;
Carlson & Boland, 1978) to replicate measured energy budget data
for the Sunset residential site in Vancouver



Issues in obtaining appropriate inputs
Q* not excessively in error but partitioning of available
energy between QH and QE very poor, primarily due to
treatment of surface water availability
Ross and Oke (1988)
The Major Themes [39]
Local-Scale Urban Energy Budget Modelling


LUMPS (Local Urban Meteorological
Parameterisation Scheme): a set of
equations, based on Q* data, and using the
OHM model for ΔQS, for partitioning the two
turbulent fluxes, based on a modification of
the de Bruin and Holtslag (1982) scheme

performs satisfactorily but requires
improvements in the database of urban
surface properties

Grimmond & Oke (2002)
TEB (Town Energy Budget) model of
Masson (2000): validation data from central
Mexico City, Vancouver light industrial site
and Marseille

performs adequately for Marseille and
Mexico City but replicates QH/ΔQS
partitioning less accurately for
Vancouver

Q* and surface temperature well
reproduced for Mexico City and
Vancouver

Masson, Grimmond & Oke (2002)

Mestayer et al. (2005)
The Major Themes [40]
Local-Scale Urban Water Budget

Grimmond, Oke & Steyn (1986)



Necessity of including anthropogenic terms in
the water budget for urban areas (piped water,
irrigation etc)
Urban areas are a mosaic of patches with very
different hydrological characteristics: irrigated
pervious (e.g. lawns - wet much of the time),
unirrigated pervious (e.g. waste ground, parks
– soil moisture varies gradually with
precipitation events) & impervious surfaces
(e.g. roads, roofs – dichotomous “wet/dry”
budgets)
Grimmond & Oke (1986)



Piped water use plays a major role in the water
budget of a Vancouver suburb (exceeds
precipitation in many months)
Possesses temporal characteristics quite
different from the other terms in the budget
(modulated by human decision making and
action, linked to geophysical events but via a
perceptual filter)
Total amounts and temporal patterns of
evaporation are consistent with the hypothesis
that irrigation of gardens (from piped water) is
the source
The Major Themes [41]
Local-Scale Urban Water Budget

Oke (1989)


Grimmond & Oke (1991)


Urban water budget is amenable to modelling incorporating formulations
adapted from forest meteorology
Grimmond & Oke (1999c)



Role of urban trees in maintaining higher rates of evaporative water loss
from cities than might be expected
While evaporation may be small in some downtown and industrial areas,
it can be 70-370% of precipitation for moist sites (and a much larger
percentage in arid areas when nearly all available water in urban areas is
piped)
Great day-to-day variability in evaporation rate, related to synoptic events
via intense coupling of the urban surface with the PBL
Richards & Oke (2002)


Suburban dew (under appropriate synoptic conditions) forms mainly in
open lawn areas and in house roofs, rather than on trees and the
roadway
Distribution of dew on heavy dewfall nights is well explained by the SVF
The Major Themes [42]
Miscellany

The UBL






Oke & East (1971)
Yap & Oke (1974)
Fuggle & Oke (1976)
Oke (1982)
Mestayer et al. (2005)
Urban Climate and Urban Planning/Design




Oke (1981)
Oke (1984)
Oke (1988a)
Oke (1989)
The Oke Contribution [1]





The Oke opus: remarkable as is the coherence, rigour and breadth of his work,
there is, in a certain sense, an anticlimactic aspect to viewing it from the vantage
point of 2006.
Why? Because novel and creative scientific contributions in an intellectually
healthy field never seem astounding in retrospect as they become incorporated
into the conceptual structure of a discipline, become routine and are adopted as
a component of the shared knowledge of the field (e.g. become taught in
courses on the topic) and so take on an element of the obvious.
Disciplines and sub-disciplines are characterised by distinctive assemblages of
conceptual structures, models, methodological procedures, terminology and
exemplars that define the intellectual culture of that area. Urban climatology is
no exception but what is perhaps exceptional is the contribution made by one
person (Oke) to the shared culture through this process of assimilation.
Indeed, a strong argument may be made for Oke as a “founding parent” of
modern urban physical climatology and as the originator of the guiding paradigm
of the field.
“Paradigm” is used here in Kuhn’s sense of disciplinary matrix rather than in his
“stronger” sense and is not meant to imply adoption of the somewhat
metaphysical elements of his philosophy that present the history of science as a
relativistic and irrational process (scientific revolutions, paradigm shifts,
incommensurability, and the like).
The Oke Contribution [2]


What are the characteristics of the Oke work
that have underpinned his role as a paradigm
builder?
Perspective will naturally reflect my values
regarding what characterizes quality, rigour
and progress in the science of urban
climatology.
The Oke Contribution [3]


Persistent emphasis on the identification of
causal processes behind urban climate
phenomena and on physical understanding
and insight.
Elevation the methodological level within
urban climate research to more effectively
extract the causal relationships and enhance
the richness of our understanding of urban
climate phenomena.
The Oke Contribution [4]
Terjung’s Levels of Methodology: Definitions
Terjung (1976)
Oke (1982)
Example
1
Qualitative
Inventory &
Associations
Recognition & description
of phenomenon
Mapping UHI & describing associations with
landuse, weather etc
2
Quantitative
Structural
Correlations
Linkage (e.g. statistically)
of the feature to other
factors
Use of regression to predict UHI ΔTu-r from wind
speed and cloudiness etc
3
Functioning
Processes
Study of processes
causing the phenomenon
Measurement of QH, flux divergence etc
4
Physical ProcessResponse Systems
Construction of processresponse models to
predict behaviour of the
phenomenon
Numerical modelling of urban SEB (e.g. TEB
model)
5
Physical-Human
Process-Response
Systems
[not included]
As above, but with two-way linkages to
social/behavioural models of humans in the city
(e.g. water and energy use)
The Oke Contribution [5]
Terjung’s Levels of Methodology: Oke’s Publications
Time Period
Level
1971-1974
2.5
1975-1979
3.3
1980-1984
3.2
1985-1989
2.9
1990-1994
3.3
1995-1999
2.6
2000-2004
2.7
2005-2006
3.2
The Oke Contribution [6]

The imperative to quantitative generalization:
to move beyond the “case study” approach of
the early days (e.g. UHI studies).





statistical generalization
mathematical modelling
hardware (scale) modelling
reduction of problems to fundamentals (e.g.
process)
framing urban climate research results in a format
suitable for users of our ideas (e.g. planners)
The Oke Contribution [7]

The goal of a predictive urban climatology.








UCL UHI intensity magnitudes
canyon energy budget and climate characteristics
local scale urban SEBs
local scale water urban budgets
urban landscape aerodynamic properties
eddy flux source areas in urban terrain
nature of a remote sensor’s sampling of urban
topography
heat storage/release at the local scale
The Oke Contribution [8]

Strength of the research programme I

Motivation: asking questions


Asking simple questions coupled intimately to a carefully
formulated program of measurements focussed on
answering the question
Asking important questions: ones to which answer will
provide insight on fundamental processes operating in
urban environment
The Oke Contribution [9]

Strength of the research programme II
 Substantive contributions to the theoretical and conceptual
framework of urban climatology





the UCL/UBL distinction (functioning as well as scale)
the distinction between different types of UHIs
the canyon as a repeating structure of the UCL
the role of the rural in defining “urban climate”
application of forest meteorology concepts in novel ways within urban
climate








the concept of the canopy
tower-based measurement techniques for areally-averaged fluxes
methods to estimate heat storage flux
precipitation interception
evaporation from sparse heterogeneous systems
roughness layer depth over rough surfaces
synoptic scale advection effects)
the importance of scale (again, not just size but process!)
The Oke Contribution [10]

Strength of the research programme III

Coherence of the body of work

The citation matrix
The Oke Contribution [11]
Oke & East (1971)
Oke & Fuggle (1972)
B
Oke (1973)
C C
Yap & Oke (1974)
B
B
Oke & Maxw ell (1975)
C C C
Fuggle & Oke (1976)
B
B
Nunez & Oke (1976)
C
C C
C
Oke (1976)
U U U U U
U
Nunez & Oke (1977)
C
C C
Oke (1979)
C
C
Nunez & Oke (1980)
C
C C C
Kalanda, Oke & Spittlehouse (1980)
B
B B
B
Oke, Kalanda & Steyn (1980/1)
B
B
Oke (1981)
C
C
C
C C
Oke (1982)
U U
U U U U U U U
Oke & McCaughey (1983)
B
B
B
Oke (1984)
Cleugh & Oke (1986)
B
B
B
Grimmond, Oke & Steyn (1986)
Grimmond & Oke (1986)
B
Oke & Cleugh (1987)
B
B
Oke (1988a) - PLAN
Ross & Oke (1988)
Oke (1988b) - PIPG
B
B
B B B
Nakamura & Oke (1988)
C
C C C
Roth, Oke & Steyn (1989)
Oke (1989)
Roth, Oke & Emery (1989)
B
Oke, Cleugh, Grimmond, Schmid & Roth (1989)
Schmid, Cleugh, Grimmond & Oke (1989)
B
Oke, Taesler & Olsson (1990)
Voogt & Oke (1991)
C
C
Johnson, Oke, Lyons, Steyn, Watson & Voogt (1991)
C
Oke, Johnson, Steyn & Watson (1991)
C C
C
Grimmond & Oke (1991)
Grimmond, Cleugh & Oke (1991)
Oke, Zeuner & Jauregui (1992)
Roth & Oke (1993)
Grimmond, Oke & Cleugh (1993)
Roth & Oke (1994)
Oke (1995)
Grimmond & Oke (1995)
Roth & Oke (1995)
Voogt & Oke (1997)
Spronken-Smith & Oke (1998)
C
C
Voogt & Oke (1998a) GEOM
Grimmond, King, Roth & Oke (1998)
Voogt & Oke (1998b) WALLS
Grimmond & Oke (1999a) - SHF
Oke, Spronken-Smith, Jauregui & Grimmond (1999)
Grimmond & Oke (1999b) -z0
Grimmond & Oke (1999c) evap
Spronken-Smith & Oke (1999)
Kanda, Moriw aki, Suzuki, Roth &Oke (2000a)
Kanda, Moriw aki, Suzuki, Roth &Oke (2000b)
Spronken-Smith, Oke & Low ry (2000)
C
Runnalls & Oke (2000)
C
C
C
Grimmond & Oke (2002)
Masson, Grimmond & Oke (2002)
Kanda, Moriw aki, Roth & Oke (2002)
Richards & Oke (2002)
Voogt & Oke (2003)
C
Offerle, Grimmond & Oke (2003)
Soux, Voogt & Oke (2004)
Grimmond, Salmond, Oke, Offerle & Lemonsu (2004)
Rotach et al. (2005)
U
Mestayer et al. (2005)
Salmond, Oke, Grimmond, Roberts & Offerle (2005)
Lagouarde, Irvine, Bonnefond, Grimmond, Oke, Salmond & Offerle (2005)
Offerle, Grimmond, Fortuniak, Klysik & Oke (2005)
Oke (2006)
C
C
B
B B
C
C
U U U U
B B
B
B B
B B
B
B B B
C
C
B B B
B B B B
C
B
B B
B B
B B
B
B
B B
B
B
C
C
B B
B B
B
B B
B
B B
B
C
C
B
C
B
B B
B
B
B
B
B B
C
C C
C C
B
B
C
C
C
B
B
B
B
B
B
B
B
B B
B
B B
B
B
B
B
B
B
B B
B
B
B
B
B
C
B
B
B
B
B
C
B
B
B
B
B
B
B B
B B
B
C
B B
B
B
C C
C
C
C
B
B
B B B
B B
B
B B
C
C
C C
C
C
B B
B B
B B
B
C
C
B
C
B
B
B
B
C
B
B
C
B B
B
B
C C
C
B
C
B
C
B
B
B
B
B
B
C
C
B B
B
B
C
C
B
B
B
C
C C
C
C
C
U
U
U
U
B
U
B
C
B
C
B B
B
U
B
C
U
B
U
U
B
B B
U
U
U U
U
U
B
B
B
C
B
C
C
U
U
B B B
B
C
C
B B B B
B B B
B
B
B
B
C
B
U
B
B
B
C
B
C C
B
B
C
C
B
C
B
B
C C
B
B
B
C
B
C
C C
B
B
B B
B
B
C
C
B
B
B
C
B
B
B
U
B
B
B
B
U
B
B B
U
The Oke Contribution [10]

Strength of the research programme III

Coherence of the body of work




Citation matrix
Persistent of themes identified in 1968-1969 McGill
Climatological Bulletin articles
Explicit posing of research questions for future work
Importance of the need for high quality
observations: sensor error analysis, instrument
exposure, time and space sampling issues etc.
The Oke Contribution [11]

Beyond the science – subdisciplinary “citizenship”




Broad scope of understanding and encouragement of
diverse approaches (e.g. collaboration between
observational and modelling scientists, particularly
emphasis on appropriate forms of model validation)
“Ambassador” for urban climatology to cognate disciplines
(e.g. urban planning/design)
Mentoring (both students and colleagues)
Seeking to upgrade and enhance the status and content of,
and internal communications within, urban climatology (e.g.
Oke, 2006)
Postscript


“The central objective of my research is to elucidate the physical basis of the
human impacts on climate caused by urban development. It requires
understanding the exchanges and balances of heat, mass and momentum as
they are modified from those of the pre-urban landscape to their present state.
This means working across a range of atmospheric scales (from micro- to
mesoscale) and relating the modification of these processes to the climatic
effects they evoke.”
(T.R. Oke, http://www.geog.ubc.ca/~toke/ResearchProjects.htm)
Download